Add Byte Offsets to the XREF Table. RES-713

Author: michaeldemey

src/main/java/com/itextpdf/rups/controller/PdfReaderController.java

@@ -44,6 +44,7 @@ This file is part of the iText (R) project.
 
 import com.itextpdf.kernel.pdf.PdfArray;
 import com.itextpdf.kernel.pdf.PdfDictionary;
+import com.itextpdf.kernel.pdf.PdfIndirectReference;
 import com.itextpdf.kernel.pdf.PdfName;
 import com.itextpdf.kernel.pdf.PdfObject;
 import com.itextpdf.kernel.pdf.PdfStream;
@@ -55,6 +56,7 @@ This file is part of the iText (R) project.
 import com.itextpdf.kernel.utils.objectpathitems.LocalPathItem;
 import com.itextpdf.kernel.utils.objectpathitems.ObjectPath;
 import com.itextpdf.rups.io.listeners.PdfTreeNavigationListener;
+import com.itextpdf.rups.model.IndirectObjectFactory;
 import com.itextpdf.rups.model.ObjectLoader;
 import com.itextpdf.rups.model.PdfSyntaxParser;
 import com.itextpdf.rups.model.TreeNodeFactory;
@@ -75,12 +77,16 @@ This file is part of the iText (R) project.
 import com.itextpdf.rups.view.itext.StructureTree;
 import com.itextpdf.rups.view.itext.SyntaxHighlightedStreamPane;
 import com.itextpdf.rups.view.itext.XRefTable;
+import com.itextpdf.rups.view.itext.treenodes.ObjectStreamTreeNode;
 import com.itextpdf.rups.view.itext.treenodes.PdfObjectTreeNode;
 
 import java.awt.Color;
 import java.awt.event.KeyListener;
 import java.util.ArrayDeque;
 import java.util.Deque;
+import java.util.List;
+import java.util.Observable;
+import java.util.Observer;
 import java.util.Stack;
 import java.util.function.Consumer;
 import javax.swing.JPanel;

src/main/java/com/itextpdf/rups/model/IndirectObjectFactory.java

@@ -46,15 +46,18 @@ This file is part of the iText (R) project.
 import com.itextpdf.kernel.exceptions.PdfException;
 import com.itextpdf.kernel.pdf.PdfDictionary;
 import com.itextpdf.kernel.pdf.PdfDocument;
+import com.itextpdf.kernel.pdf.PdfIndirectReference;
 import com.itextpdf.kernel.pdf.PdfName;
 import com.itextpdf.kernel.pdf.PdfNull;
 import com.itextpdf.kernel.pdf.PdfObject;
+import com.itextpdf.kernel.pdf.PdfStream;
 import com.itextpdf.rups.view.Language;
 
 import java.lang.reflect.Field;
 import java.lang.reflect.InvocationTargetException;
 import java.lang.reflect.Method;
 import java.util.ArrayList;
+import java.util.List;
 
 /**
  * A factory that can produce all the indirect objects in a PDF file.
@@ -77,6 +80,11 @@ public class IndirectObjectFactory {
      * A list of all the indirect objects in a PDF file.
      */
     protected ArrayList<PdfObject> objects = new ArrayList<>();
+
+    /**
+     * List of all Object Streams in a PDF file.
+     */
+    protected List<PdfIndirectReference> objectStreams = new ArrayList<>();
     /**
      * Mapping between the index in the objects list and the reference number in the xref table.
      */
@@ -145,6 +153,14 @@ public boolean storeNextObject() {
                 final int idx = size();
                 idxToRef.put(idx, current);
                 refToIdx.put(current, idx);
+
+                if ( object.getType() == PdfObject.STREAM ) {
+                    PdfStream stream = (PdfStream) object;
+                    if ( PdfName.ObjStm.equals(stream.get(PdfName.Type) )) {
+                        this.objectStreams.add(stream.getIndirectReference());
+                    }
+                }
+
                 store(object);
                 return true;
             }
@@ -238,6 +254,10 @@ public boolean isLoadedByReference(int ref) {
         return isLoaded.get(getIndexByRef(ref));
     }
 
+    public List<PdfIndirectReference> getObjectStreams() {
+        return objectStreams;
+    }
+
     /**
      * Loads an object based on its reference number in the xref table.
      *

src/main/java/com/itextpdf/rups/model/ObjectLoader.java

@@ -120,6 +120,7 @@ public TreeNodeFactory getNodes() {
         return nodes;
     }
 
+
     /**
      * getter for a human readable name representing this loader
      *

src/main/java/com/itextpdf/rups/view/Language.java

@@ -201,6 +201,7 @@ public enum Language {
     PAGES,
     PAGES_TABLE_OBJECT,
     PDF_READING,
+    PDF_OBJECT_STREAMS_TREE_NODE,
     PDF_OBJECT_TREE,
     PLAINTEXT,
     PLAINTEXT_DESCRIPTION,
@@ -246,10 +247,15 @@ public enum Language {
     WARNING_OID_NAME_NOT_FOUND,
 
     XREF,
+    XREF_BYTE_OFFSET,
+    XREF_BYTE_OFFSET_OBJECT_STREAM,
     XREF_DESCRIPTION,
+    XREF_NA,
+    XREF_NOT_LOADED_YET,
     XREF_NUMBER,
     XREF_OBJECT,
-    XREF_READING;
+    XREF_READING
+    ;
 
     /**
      * The location of the resource bundles.

src/main/java/com/itextpdf/rups/view/itext/ObjectStreamParser.java

@@ -0,0 +1,55 @@
+package com.itextpdf.rups.view.itext;
+
+import com.itextpdf.io.source.PdfTokenizer;
+import com.itextpdf.io.source.RandomAccessFileOrArray;
+import com.itextpdf.io.source.RandomAccessSourceFactory;
+import com.itextpdf.kernel.pdf.PdfName;
+import com.itextpdf.kernel.pdf.PdfStream;
+import com.itextpdf.rups.model.LoggerHelper;
+import com.itextpdf.rups.view.Console;
+
+import java.io.IOException;
+import java.util.Arrays;
+
+/**
+ * Utility class to parse ObjectStreams to extract the offset of a given object id within the stream.
+ */
+public class ObjectStreamParser {
+    /**
+     * Parses an ObjectStream to find the offset to the passed parameter, compressedObjectNumber. This offset is
+     * relative to the ObjectStream and not to the complete file, as described in the specification.
+     *
+     * If an Object ID isn't found inside the ObjectStream, this will return -1.
+     *
+     * @param objStm The ObjectStream to parse
+     * @param compressedObjectNumber the ID of the object of which you want the offset
+     * @return the offset of the object or -1 if the object is not found
+     */
+    public static int parseObjectStream(PdfStream objStm, int compressedObjectNumber) {
+        byte[] objStmBytes = objStm.getBytes(true);
+        int byteOffsetOfFirst = objStm.getAsInt(PdfName.First);
+
+        PdfTokenizer pdfTokenizer = new PdfTokenizer(
+                new RandomAccessFileOrArray(
+                        new RandomAccessSourceFactory()
+                                .createSource(
+                                        Arrays.copyOfRange(objStmBytes, 0, byteOffsetOfFirst))));
+
+        try {
+            while (pdfTokenizer.nextToken()) {
+                if ( pdfTokenizer.getTokenType().equals(PdfTokenizer.TokenType.Number )) {
+                    int objNumber = pdfTokenizer.getIntValue();
+                    pdfTokenizer.nextToken();
+                    if ( objNumber == compressedObjectNumber ) {
+                        return pdfTokenizer.getIntValue();
+                    }
+                }
+            }
+        } catch (IOException e) {
+            LoggerHelper.error(e.getMessage(), e, ObjectStreamParser.class);
+            return -1;
+        }
+
+        return -1;
+    }
+}

src/main/java/com/itextpdf/rups/view/itext/XRefTable.java

@@ -42,8 +42,7 @@ This file is part of the iText (R) project.
  */
 package com.itextpdf.rups.view.itext;
 
-import com.itextpdf.kernel.pdf.PdfNull;
-import com.itextpdf.kernel.pdf.PdfObject;
+import com.itextpdf.kernel.pdf.*;
 import com.itextpdf.rups.controller.PdfReaderController;
 import com.itextpdf.rups.model.IndirectObjectFactory;
 import com.itextpdf.rups.model.ObjectLoader;
@@ -93,7 +92,7 @@ public XRefTable(PdfReaderController controller) {
      */
     @Override
     public int getColumnCount() {
-        return 2;
+        return 3;
     }
 
     /**
@@ -117,6 +116,8 @@ public Object getValueAt(int rowIndex, int columnIndex) {
                 return getObjectReferenceByRow(rowIndex);
             case OBJECT_COLUMN_INDEX:
                 return getObjectDescriptionByRow(rowIndex);
+            case 2:
+                return getByteOffSetByRow(rowIndex);
             default:
                 return null;
         }
@@ -147,6 +148,47 @@ private String getObjectDescriptionByRow(int rowIndex) {
         return PdfObjectTreeNode.getCaption(object);
     }
 
+    /**
+     * Returns the byte offset of the selected XREF entry. If the entry has no real, actual offset.
+     * i.e. it is compressed in a PDF Object Stream, then this shall return The ID of the Object Stream
+     * and the offset of the Object Stream.
+     *
+     * @param rowIndex the index of the selected XREF entry
+     * @return byte offset of the XREF entry or the ID and byte offset of the encompassing Object Stream
+     */
+    private String getByteOffSetByRow(int rowIndex) {
+        final PdfObject object = objects.getObjectByIndex(rowIndex);
+        PdfIndirectReference indirectReference = object.getIndirectReference();
+        if ( indirectReference != null ) {
+            if (isObjectStream(indirectReference)) {
+                int compressedObjectNumber = indirectReference.getObjNumber();
+                PdfStream objStm = getObjectStream(indirectReference);
+                int internalCompressedObjectOffset
+                        = ObjectStreamParser.parseObjectStream(objStm, compressedObjectNumber);
+
+                return String.format(
+                        Language.XREF_BYTE_OFFSET_OBJECT_STREAM.getString(),
+                        objStm.getIndirectReference().getObjNumber(), internalCompressedObjectOffset
+                );
+            }
+
+            return String.valueOf(indirectReference.getOffset());
+        }
+        return Language.XREF_NOT_LOADED_YET.getString();
+    }
+
+    private PdfStream getObjectStream(PdfIndirectReference indirectReference) {
+        int objStreamNumber = indirectReference.getObjStreamNumber();
+        PdfObject objectByIndex = objects.loadObjectByReference(objStreamNumber);
+
+        return (PdfStream) objectByIndex;
+    }
+
+    private boolean isObjectStream(PdfIndirectReference indirectReference) {
+        return indirectReference.getOffset() == -1;
+    }
+
+
     /**
      * @see javax.swing.JTable#getColumnName(int)
      */
@@ -157,6 +199,8 @@ public String getColumnName(int columnIndex) {
                 return Language.XREF_NUMBER.getString();
             case OBJECT_COLUMN_INDEX:
                 return Language.XREF_OBJECT.getString();
+            case 2:
+                return Language.XREF_BYTE_OFFSET.getString();
             default:
                 return null;
         }

src/main/java/com/itextpdf/rups/view/itext/treenodes/ObjectStreamTreeNode.java

@@ -0,0 +1,13 @@
+package com.itextpdf.rups.view.itext.treenodes;
+
+import com.itextpdf.kernel.pdf.PdfObject;
+import com.itextpdf.rups.view.Language;
+
+public class ObjectStreamTreeNode extends PdfObjectTreeNode {
+
+    public ObjectStreamTreeNode(PdfObject object) {
+        super(object);
+        setUserObject(Language.PDF_OBJECT_STREAMS_TREE_NODE.getString());
+    }
+
+}

src/main/resources/bundles/rups-lang.properties

@@ -167,6 +167,7 @@ PAGES=Pages
 PAGES_TABLE_OBJECT=Object %d
 
 PDF_READING=Reading PDF document...
+PDF_OBJECT_STREAMS_TREE_NODE=PDF Object Streams
 PDF_OBJECT_TREE=PDF Object Tree (%s)
 
 PLAINTEXT=Plain Text
@@ -220,7 +221,11 @@ WARNING_OPENED_IN_READ_ONLY_MODE=Document is opened in a read-only mode. For edi
 WARNING_OID_NAME_NOT_FOUND=Full name for object identifier was not found: %s.
 
 XREF=XREF
+XREF_BYTE_OFFSET=Byte Offset
+XREF_BYTE_OFFSET_OBJECT_STREAM=Object Stream #%d (%d)
 XREF_DESCRIPTION=Cross-reference table
+XREF_NA=N/A
+XREF_NOT_LOADED_YET=Offset not loaded yet
 XREF_NUMBER=Number
 XREF_OBJECT=Object
 XREF_READING=Reading the Cross-Reference table

src/main/resources/bundles/rups-lang_nl_NL.properties

@@ -221,6 +221,7 @@ WARNING_OID_NAME_NOT_FOUND=Volledige naam voor object identifier niet gevonden:
 
 XREF=XREF
 XREF_DESCRIPTION=Cross-reference tabel
+XREF_NOT_LOADED_YET=Offset is nog niet ingeladen
 XREF_NUMBER=Nummer
 XREF_OBJECT=Object
 XREF_READING=Lezen van de Cross-Reference tabel